29 July 2020
Introducing the CPR COVID-19 Research Collaborative

In April 2020, the Centre for Policy Research (CPR), began a collaboration with the Government of Punjab (GoP) to revise the State’s COVID-19 testing strategy. The aim of this effort was to move beyond current strategies centred on testing symptomatic patients, towards a broader, surveillance-based testing model. The need for this shift arises from the limitations of symptomatic testing in supplying key insights about the disease. Contact tracing only symptomatic patients, once community transmission begins, does not provide an accurate understanding of prevalence rates, or how to contain the spread of the disease effectively. Because finding infected people as early as possible helps to control the epidemic through quarantine/localised lockdowns, early detection was the key objective in the model’s design. Accordingly, CPR based the strategy’s development on two hypotheses:

  • First, the prevalence of asymptomatic potential ‘super spreaders’, which made current symptomatic testing protocols insufficient;
  • Second, the spatial clustering of disease outbreaks, which made dividing the population into subgroups to identify ‘who’ to test (stratified random sampling), a necessary condition. An alternative way to do this is to establish better ‘predictors’ of disease spread, and identify people to test in a statistically structured, randomised manner.

Understanding the nature of the disease as it unfolded in Punjab, was the first step in the collaboration. The team worked closely with GoP to understand how existing protocols, data collection systems, and disease management processes were being implemented. After gathering this data, CPR undertook a detailed analysis to test the hypotheses. These research notes are based on this analysis, and published with the aim of using emergent available data to improve our understanding of COVID-19.

There is however, an important caveat to this analysis. As anyone familiar with government datasets in India, and especially those engaged with COVID-19 specific data, are well aware, there are quality issues that emerge from the fact that existing databases are poorly designed and do not lend themselves to easy data entry (or subsequent analysis). Moreover, there are serious capacity constraints on the ground, as overburdened health and other frontline workers may lead to delays and errors in data entry. We have worked hard to try and address these challenges. We believe that, despite its limitations, assessing data as it becomes available contributes valuably to our understanding of the nature of the disease, as well as the kind of policy decisions that can be taken. It is with this spirit that we present these research notes.

The views shared belong to individual faculty and researchers and do not represent an institutional stance on the issue.